Share This Article
UK banks are ramping up trials of “agentic AI”, autonomous systems that can plan, decide, and execute tasks without constant human prompting, as regulators warn the technology could introduce a new category of risks to both consumers and financial stability.
Unlike standard generative AI tools that mainly produce text or code, agentic AI is designed to take action. In banking, that could mean tools that automatically shift idle cash into better-rate accounts, adjust portfolios in response to market moves, or actively manage budgets and spending controls in real time. The promise is obvious: better personalization, faster service, and smarter financial decision-making at scale.
But the UK’s Financial Conduct Authority (FCA) has been clear that this shift raises serious operational concerns. When autonomous AI is making decisions at high speed, and potentially interacting with other automated systems, mistakes can spread faster, be harder to detect, and become harder to roll back. The regulator has flagged that these tools could amplify consumer harm if safeguards fail, especially in sensitive areas like financial advice, lending, and spending decisions.
There are also growing systemic risk concerns. One of the key warnings is that if many banks deploy similar agents trained on similar signals, they may all react in the same way at the same time, potentially accelerating volatility or even triggering “herding” behavior in stressed conditions. In extreme cases, rapid automated money movement between institutions could worsen liquidity stress and increase the risk of destabilizing events.
The FCA’s approach so far has been to apply existing frameworks rather than introduce AI-specific rules immediately, including principles like Consumer Duty and the Senior Managers regime, which places accountability on leadership. Alongside this, regulators are increasing scrutiny and beginning deeper review work on the overall impact of advanced AI on retail financial markets.
Overall, the UK banking sector appears to be moving quickly toward real-world deployment, but the regulator’s messaging is clear: as agentic AI moves from back-office experimentation to customer-facing decision-making, governance, testing, and resilience will determine whether it becomes a breakthrough for retail finance or a new source of instability.